Mamy Ratsimbazafy

mratsim's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

New activity in nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4 about 3 hours ago

NVFP4 cannot be loaded in SGLang

1

#12 opened about 3 hours ago by

mratsim

New activity in mratsim/MiniMax-M2.5-BF16-INT4-AWQ 1 day ago

accuracy

21

#4 opened 27 days ago by

ktsaou

New activity in mratsim/MiniMax-M2.5-BF16-INT4-AWQ 6 days ago

Any plans for Xiaomi's MiMo V2 Flash?

2

#8 opened 22 days ago by

droussis

Are there any plans to make BF16/FP8 AWQ INT4 version of Qwen/Qwen3.5-397B-A17B?

❤️ 1

2

#7 opened 23 days ago by

zuuky

New activity in mratsim/MiniMax-M2.5-BF16-INT4-AWQ 24 days ago

Links in README

1

#6 opened 24 days ago by

Jon-Nielsen

FP8 + INT4 version

14

#2 opened 27 days ago by

bigstorm

New activity in mratsim/MiniMax-M2.1-FP8-INT4-AWQ 26 days ago

Could you cook a similar version for Step 3.5?

👀 1

7

#8 opened about 1 month ago by

bigstorm

New version for M2.5

🔥 3

5

#10 opened 28 days ago by

blobbybob

New activity in mratsim/MiniMax-M2.5-BF16-INT4-AWQ 26 days ago

Cant get it to work on 8x RTX3090

14

#1 opened 27 days ago by

maglat

vLLM's log reports a 'incorrect tokenization'.

2

#5 opened 27 days ago by

PowAG

New activity in mratsim/MiniMax-M2.5-BF16-INT4-AWQ 27 days ago

accuracy benchmark

1

#3 opened 27 days ago by

mwalol

New activity in mratsim/MiniMax-M2.1-FP8-INT4-AWQ 27 days ago

nvfp4

12

#9 opened 29 days ago by

festr2

New activity in mratsim/MiniMax-M2.1-BF16-INT4-AWQ 27 days ago

Fastest for my 3090x8

2

#1 opened 28 days ago by

ciprianv

New activity in mratsim/MiniMax-M2.1-FP8-INT4-AWQ 28 days ago

Kind request: GLM-4.7-Flash

2

#6 opened about 2 months ago by

dehnhaide

New activity in XiaomiMiMo/MiMo-V2-Flash about 1 month ago

flashinfer报错， 是否支持 RTX 6000pro blackwell？

4

#14 opened 3 months ago by

lluu8

New activity in unsloth/Kimi-K2.5-GGUF about 1 month ago

Original model is just 600gb

14

#2 opened about 1 month ago by

jpsequeira

New activity in mratsim/GLM-4.7-Flash-FP8 about 2 months ago

Kind request

5

#1 opened about 2 months ago by

dehnhaide

New activity in baichuan-inc/Baichuan-M3-235B-GPTQ-INT4 about 2 months ago

wrong reasoning parser

2

#1 opened about 2 months ago by

mratsim

New activity in mratsim/MiniMax-M2.1-FP8-INT4-AWQ about 2 months ago

ValueError: Unsupported weight strategy=block, supported strategies are [<QuantizationStrategy.CHANNEL: 'channel'>, <QuantizationStrategy.TENSOR: 'tensor'>]

8

#5 opened about 2 months ago by

ablueleaf

New activity in baichuan-inc/Baichuan-M3-235B about 2 months ago

Suitability for medical research?

#2 opened about 2 months ago by

mratsim